The Future of Lighting Controls

BlogPost template by Adobe Dreamweaver

The Future of Lighting Controls

Disclaimer: The Integrated Design Lab recently had LLLCs installed in part of our office and actively promote this emerging lighting technology. Additionally, the IDL reviews Idaho Power’s new construction verification program which added a new incentive for LLLCs in the 2019 version.

The purpose of building energy codes

Building energy codes and standards set minimum requirements for energy efficient design and construction of new buildings as well as additions that impact energy use and emissions for the lifespan of the building. Standards or codes impact the market by creating demand that must be met with new practices or technologies from the built environment and its practitioners.

The practice of lighting design has made a resurgence in recent decades where both daylight and electric light are considered for buildings and space types. In addition, returning to an old practice there have been significant efforts to modernize the practice of lighting design as a science (Lighting Levels, LPD values, and Glare Index).

Where are we going?

The standards regarding lighting controls have evolved significantly over the past three versions of IECC 2012, 2015, and 2018. Each version that succeeded the last has required more types of control as well as requiring more space types to have controls. Of these control strategies the common request for more information is for the daylight zoning and controls.

Daylight Zone Control C405.2.2.3 (IECC 2012)

  • Lights in the daylight zone are controlled independently of general area lighting and are controlled in accordance with either Manual or Automatic daylighting controls.
  • Daylight zone control shall not exceed 2,500 square feet.
  • Continuous daylight zones adjacent to vertical fenestration are allowed to be controlled by a single controller provided that they do not include zones facing more than two adjacent cardinal orientations.
  • Daylight zones under skylights more than 15 feet from the perimeter shall be controlled separately.
  • 1 Exception.
  • Automatic Daylighting controls shall be capable of automatically reducing the lighting power in response to available daylight by either one of the following methods; Continuous or Stepped Dimming.
  • New section added C405.2.3.1 Daylight-responsive control function has six standards about controls, dimming, space types, and zone control in relation to cardinal orientation.

    New sections to define Side-Lighting and Top-Lighting (Windows, Rooftop monitors, Clerestories, and Skylights) which also defines how to calculate a daylight area. Clarifies properties of fenestration types such as minimum area required, visible transmittance, and exterior shading. One value that should be considered is the window top sill height in relation to the ground plane.

    Green text is new whereas red is text that was removed from code and black text has remained the same.

    Other changes worth noting:

    C405.3.1 = Total connected interior lighting power

    C405.2.3.1 = Daylight-responsive control function unchanged

    C405.2.3.2 = Changed to Sidelit Zone and rule two moved into its own section of C405.2.3.3 Toplit Zone which previously only dealt with skylights.

    C405.2.6 = Exterior Lighting Controls now require compliance with daylight shutoff, lighting setback (reduction of power), or time switch control function.

    New section C405.2.5 Manual controls - While previously taken out along with requirements of automatic controls they were reintroduced to avoid confusion from different lighting requirements. Below are figures 1 -3 that demonstrate how to draw the daylight zone as defined by IECC 2018.

    Figure 1 - Sidelit Daylight Zone

    Figure 2 - Rooftop Monitor Daylight Zone

    Figure 3 - Roof Fenestration Assembly Daylight Zone

    So as noted above, standards and code requirements are evolving so that it is no longer enough to have lighting in a large area operate with ON or OFF. Lighting must adapt to individual needs in response to changing occupancy, daylight, desired light level, and external energy grid conditions. The following are the top emerging lighting controls strategies.

    Occupancy Sensors

  • Sensors that automatically turn off or dim lights when a person leaves a room or space.

  • Scheduling

  • Lights are automatically dimmed or turned off at certain times of the day.

  • Multi-Level Lighting or Dimming

  • Control devices that adjust lighting power by continuous dimming, stepped dimming, or stepped switching.

  • Building Automation System (BAS)

  • BAS are most often used to control the mechanical, electrical, and plumbing systems in buildings, but they can also be used provided lighting control based on occupancy scheduling.

  • Daylight Harvesting

  • A lighting control and shade system that automatically dims light fixtures when natural light is available, and/or adjusts shades so that natural light and artificial light combine to provide the desired level of lighting.

  • Demand-Response Lighting

  • Reduces lighting by dimming or turning off lights at times of peak electricity pricing.

  • Plug Load Control

  • Device that automatically turns off wall plugs and lights when a person leaves a room or space.

  • As a direct result of stricter code requirements and emerging control strategies the market has responded to these demands with the creation of Luminaire Level Lighting Controls or LLLCs. LLLCs are intended to be a one stop shop so that they meet or exceed lighting requirements in all building types for any standard. One goal for LLLCs from manufacturers is to achieve a plug and play level of simplicity where systems require little to no additional programming cost after installation. The intention of this goal is to be used for offsetting the initial increased cost of LLLC fixtures.

    What are LLLCs?

    Luminaire Level Lighting Controls, LLLCs, unlike most acronyms in our industry, is fairly straight forward in that it is lighting controls at the luminaire level or the individual fixture. For a fixture to qualify as an LLLC it must have the capability to have a networked occupancy sensor and ambient light sensor installed for each luminaire, and directly integrated or embedded into the luminaire form factor during the luminaire manufacturing process. In addition to these required integrated components, LLLC systems must have control persistence capability. To demonstrate commercial availability of the integrated component options, at least one family, luminaire or kit with integrated control must be verified by Design Light Consortium (DLC).

    Figure 4 - Example of integrated sensor for an LLLC fixture

    Figure 5 - Example of an LLLC fixture

    The Design Lights Consortium (DLC) is a non-profit organization dedicated to accelerating the widespread adoption of high-performing commercial lighting solutions. The DLC promotes high-quality, energy-efficient lighting products in collaboration with utilities and energy efficiency program members, manufacturers, lighting designers, and federal, state, and local entities.

    The IECC defines LLLC as a system in which luminaries feature embedded intelligence, occupancy and light sensors, wireless networking capability, and where required local override switching capability. The code requires the luminaire to be independently capable of occupancy sensing, dimming to maintain a desired light level, and configurability including dimming set-points, timeouts, fade rates, sensor sensitivity, and wireless zoning.

    C405.2 Lighting Controls (Mandatory)

  • Lighting controls as specified in Sections C405.2.1 through C405.2.6
  • The LLLC luminaire shall be independently capable of the following:
  • Monitoring occupant activity to brighten or dim lighting when occupied or unoccupied, respectively.
  • Monitoring ambient light, both electric and daylight, and brighten or dim artificial light to maintain desired light level.
  • For each control strategy, configuration and reconfiguration or performance parameters including; bright and dim setpoints, timeouts, dimming fade rates, sensor sensitivity adjustments, and wireless zoning configurations.

  • For simplicity sake, it means the following:

    Figure 6

    Figure 7

    Types of LLLCs

    Due to the wide range of applications for this technology there are three tiers for classifying your type of LLLCs. The different tiers associate the level of sophistication and intelligence the system is capable of achieving. The three tiers are ‘clever’, ‘clever-hybrid’, and ‘smart’ systems as defined by the Northwest Energy Efficiency Alliance or NEEA. Clever systems are classified as LLLCs which meet basic DLC qualified products list. The basic requirements are requirements for the lights that must be possible but it doesn’t mean you will necessary need all of the control strategies available (high-end trim, dimming, occupancy, and photocontrols). A smart system includes all the clever system capabilities but in addition is able to communicate energy and non-energy data. Building owners or facilities managers can use this data to inform their decision making processes for a variety of applications, such as, HVAC optimization, asset tracking, or Internet of Things (IoT). The clever-hybrid system falls in between where they have all of the clever capabilities but lack the capacity to be considered a smart system or full IoT capabilities. Clever-hybrid systems typically involve an independent network gateway that allows some data management but is usually energy-use data specific.

    Figure 8

    Built Environment's IoT

    IoT as defined by is the interconnection via the Internet of computing devices embedded in everyday objects, enabling them to send and receive data. The internet of things can be considered a ‘sensor network’ in which they are connected to the internet. However, these ‘things’ for our intended purpose are thermostats, occupancy sensors, photocells, mobile devices, security locks, plug-loads, etc.… When ‘things’ are connected to a network they can create a ‘sensor network’ that can be used to solve energy management in buildings as well as assisting future planning/decision making.

    Figure 9

    Figure 10

    Figure 9, above, shows the vast reach of the IoT, however, our focus is its applicability to our buildings which is shown in figure 10. IoT has three main characteristics, obtaining information, reliable delivery, and processing data for analysis.

    IoL - Internet of Lighting

    “An Internet of Things (IoT) indoor lighting system that consists of multiple luminaries with an IP address, sensors and controllers is considered. The sensors provide sensing information for controlling artificial lighting system and additionally serve as a data source for other building systems and services.” (Warmerdam 1) In systems that account for personal control, occupants can change the illumination in their surroundings by communication using an app, software interface, or a web-based application to the lighting system. The uses for this include integration of intelligent technology for optimization of building energy systems or intelligent system of unified planning, design, equipment, and systems to achieve resource sharing with information exchange.

    Lighting systems are evolving to support different wired/wireless communications interfaces compatible with the IoT ecosystem. The overall goal is to join drivers with sensors so that algorithms and wireless communications are able to deploy scalable lighting solutions capable to work autonomously in the IoT ecosystem. Therefore, smart lighting, figure 10, should be conceived as an adaptable lighting system with the objective to improve visual or occupant comfort first then energy efficiency.

    Zone vs Area Control

    In conventional lighting systems lighting zones are defined as a collective unit and are centrally controlled where as luminaire level controls has the potential to become a semi-autonomous zone that is capable of responding to small changes in the area under each fixture. The fixture based approach increases the quantity of sensors which results in more data points available to record inputs from occupants and space conditions. One way to think about the differences between the two approaches is to consider dpi, dots per inch. Where dots in this instance are the sensors available to collect data. In the zone based approach with only a handful of sensors in a space we can expect to get an image resolution of 72dpi. Now if we consider the fixture based approach where we have significantly more sensors or dots available to gather data which allows us to achieve a higher dpi, such as, 300dpi.

    Figure 11 - Zoning area vs fixture based approach

    The amount of sensors available in a space determines the amount information available to be analyzed which allows facilities managers to make informed decisions, however, there such a thing as too much data. Most buildings or space types do not need to have a sensor for each light fixture collecting data in order to achieve optimal control strategies or provide relevant feedback. As an Architect, Engineer, or designer you need to be able to determine the correct amount of lights needed for a space as well as the number of sensors, switches, or access points for your control strategies to be effective or utilized by the building’s occupants. The same concept is applicable for LLLCs in a space where generally speaking not every fixture needs to have embedded sensors for control strategies to be effective. However, if all lighting in a building is LLLCs then multiple new avenues of control and communication can be achieved as well as being able to scale the control strategies to fit the building’s needs.

    Occupancy sensors with large coverage patterns can commonly translates into a circle with thirty to fifty feet diameter depending on manufacturer and model (2,000 square feet in coverage area). Your typical office employee has a limited activity area, i.e. a cubicle that is ten feet by ten feet would roughly translate to one-hundred square feet per employee. In this situation each sensor could respond to movements or occupancy of up to twenty different people. Also, if a space only has a single occupancy sensor then all the fixtures will respond to a single input regardless of occupancy levels. One study, General Services Administration (GSA) study, indicates that the typical office worker is away from their desk half of the time, if this vacancy can be captured, then energy savings can be maximized. “22 of 82 workstations have less than three hours of occupied time on at least half the days studied” (Enscoe 4). The Zone based approach tends to overlap into other spaces due to its wide range of coverage which can mitigate the amount of energy savings achieved in a building or space.

    For Example

    So lets say we have 6 cubicles on the north and south side of the space with a hallway splitting them. In this space there are two open work area for team members to collaborate and occasionally the spaces are combined for meetings or presentations.

    Figure 12

    Here, Figure 13, is how the manufacturer intends for the control strategies to be implemented. However, this is not an environment people are accustomed to working in, therefore I would recommend a bit of coloring outside the lines.

    Figure 13

    So we can change the programming to instead of full off when no occupancy is detected during normal work hours to be 40% of full on.

    Figure 14

    We know that when someone enters the Open Work Area they are usually not alone and are going to occupy a majority of the space, therefore, we want to group those lights.

    Figure 15

    So when four team members leave their work area or cubicles.

    Figure 16

    The hallway lights occupancy sensor is triggered and the lights turn on full.

    Figure 17

    Then when they enter the open work area all four lights are turned to full on rather than one at a time when their occupancy sensor is triggered.

    Figure 18

    Now lets say there are some windows that allow natural daylight into the open work areas. In that case lights would dim in response to the amount of available daylight according to its photocell.

    Figure 19

    Figure 19A

    One factor to note, again, is occupant comfort and how difficult it is to derive a scale for measuring lighting comfort or Glare Index to individuals. LLLCs offer flexible programing to the degree that each occupant can have the fixture output level match their standard for comfort. Increased granularity may help to reduce energy similar to occupancy sensors, but more importantly it may also help to provide greater uniformity of target illumination.

    Figure 20

    Selecting a LLLC product and settings for maximum energy savings has the potential to adversely impact occupant satisfaction, therefore, energy efficiency but not at the cost of occupant comfort.

    Energy Efficiency

    Figure 22

    Different control strategies will yield different rates of savings but their overall goal is to reduce energy consumption while promoting occupancy comfort. One of the highest savings for lighting systems is though upgrading fixtures from compact fluorescent to LEDs. Upgrading fixtures comes with a fixed savings cost that can be calculated whereas control strategies savings are expressed in ranges because of it’s dependency on how occupants use the space or conditions throughout the day, month, and year. In addition, lighting control strategies, much like lighting design, needs to be applied in layers as each strategy will impact the overall performance or use of lighting in a space. The ranges from control strategies or parameters that affect lighting design are the following:

  • Field of view for the built-in motion sensor.
  • The delay time between when the occupancy is detected and when the luminaire is turned off or dimmed.
  • Number of luminaires that turn on and off together in groups.
  • If the luminaires dim to a low level or turn off completely when no motion is detected.

  • For motion sensors the wider the angle covered by the motion sensor, the lower the expected energy savings because more motion events will trigger the luminaire to turn on. Delay times for occupancy sensors guard against a false negative. For instance, a longer delay time means a higher chance the occupant will make a sufficiently large motion before the light is turned off, thereby reducing the chance of annoying the occupant. However, when the monitored area is actually unoccupied, the delay time results in wasted light or failed capture of energy savings.

    Another option to consider when no motion is detected under a luminaire is to have it dim to a low level rather than turn completely off. In addition, control strategies can be configured to have a double trigger for last occupancy where the first trigger will dim the lights to twenty percent, for example, and then if the second trigger is allowed to pass then the lights will turn off. This two-step trigger will provide occupants in a space with a visual notification that the lights will be turning off in X minutes and if they want the light to stay on then they can trip the motion sensor or communicate with the lights via a software application. Considering this option further we can use it to provide a sense of normalcy in space, depending on the size. Most people are not comfortable or rather not used to working in a open space where the only lighting on in a space is where demand is being actively tracked or triggered.

    The last significant parameter to consider is grouping or as mentioned earlier zone vs. fixture base approach. LLLC products can form mesh networks, by which, luminaries can then be programmed into groups so that if motion is detected by one luminaire, then all luminaries in the group will turn or remain on.

    Occupants may wish to group luminaries for aesthetic reasons or to identify a cohesive work department. However, the greater the number of luminaries grouped together will eventually be equal to the number of luminaires grouped by traditional controls. Choices made during the LLLC selection (i.e. sensor field of view) and commissioning (i.e. delay time, grouping, dimming vs full off for vacancy) have an impact on the potential energy savings through operation.

    Cost-benefit analysis

    There are two avenues of cost that need to be explored when considering LLLCs, per fixture cost and parameter settings. The table below demonstrates the change in fixture cost over four years, from a NEEA study, which shows an overall decrease in cost for all three tiers of LLLCs. The change is cost for fixtures can be associated with competition, economies of scale, and improved manufacturing methods. These cost estimates do not include cost of commissioning, software subscriptions, or any programing set-up required for building occupants to utilize the lighting controls. If a building is being retrofitted or entirely re-designed from compact fluorescent to LED fixtures the argument for choosing LLLCs as the replacement is considerably easier because of the cost-benefit analysis can be justified by the amount of kWh saved. Whereas, a similar argument can be made with photocontrols for a building in that if there is 100 kWh pool of energy from compact fluorescent against a 50 kWh energy pool from LEDs and your photocontrol strategy will save 50% of your kWh. In this scenario you would save more energy, with photocontrols, if you were using a compact fluorescent fixtures rather than LED fixtures. Therefore, you need to consider the context of control parameters and understand that depending on that context they will perform better or worse. Context includes but not limited to building type, building use, occupants, all fixtures, and other control strategies.

    “2020 Luminaire Level Lighting Controls Incremental Cost Study.” NEEA, January 7, 2021.

    As we are all aware the creation of a design and the implementation of that design are two very different entities. The following tables shows the system cost components, per fixture cost, and per square foot cost assuming a 40,000 square foot building.

    In comparison we see the following:

    Figure 23

    We are only going to consider the following parameters for this blog as they are the most commonly used or required, Occupancy/Vacancy sensors, Photocontrols, Scheduling, Dimming, and Occupant overrides. Some examples of how control strategies context will effect the impact of energy efficiency are the following:

  • A delay time of 5 minutes vs a typical 20 minute delay reduces energy use by 21%.
  • Leaving luminaires ungrouped reduces energy use up to 29% when compared to connecting luminaires into groups of 8.
  • A narrow field of view sensor selection vs a wide field of view sensor will reduce energy use by 18%.
  • Turning off luminaires during vacancy reduces energy use up to 14% compared with dimming troffers to 20% light output when vacancy is detected under the troffer but there is occupancy elsewhere in the room.

  • These numbers are from a study, “Saving Energy with Highly-Controlled Lighting in an Open-Plan Office.”, where the focus was to identify the impacts control strategies will have on energy savings depending on the parameters selected for implementing lighting control strategies. The parameters that were evaluated are delay time, grouping fixtures, final occupancy, and field of view. These parameters are among the most commonly seen in commercial buildings of various types. In fact, building occupants generally expect these parameters to be present in commercial buildings where these control strategies help define the type of environment for a space.

    While we are limiting our scope it is important to note that evaluating multiple control parameter settings can lead to a large amount of data to analyze. There is a fundamental pattern that lighting controls follow that revolves around trigger events. A trigger event is an event that triggers a change in the lighting behavior for a single space or the entire building. Either the lighting schedule or the occupancy sensor will cause a trigger event for the lights to turn on and will stay on until another trigger event occurs. Typically, the lighting in a building will turn on at a designated morning time, 8am let’s say, and will stay on until the last occupancy event is detected where the lights for a space will turn off. This feature helps refine the lighting schedule event which turned on all of the lights. The cycle of events will continue with a if/then check performed each time until the conditions are met. The if/then checks are configured for the lighting schedule so that ‘IF’ the time of day is X:XX, 8pm let’s say, then turn off the lights or leave the lights on. At that point the lights will turn off in response to the if/then trigger event with the option of the occupancy sensor to override the command.

    Figure 24

    So, with that basic fundamental pattern we can infer a significant amount of data will be produced by analyzing the potential settings we can use for our building’s lighting control parameters. Now if we were to consider the parameter settings available for LLLCs using the figures below, 25-29, you can start to see that the amount of data to analyze is significant.

    Figure 25 – The chart starts out with ‘Lights On’ and then branches to two options of controls present or not present. We will select controls for this example and have our schedule turn on all lights at a determined time.

    Figure 26 – Since we are looking at LLLCs we will select our level of control to be at the fixture level. We’ll also choose occupancy sensor since our fixture will have them embedded.

    Figure 27 – Now we have to select our parameters for our occupancy sensor which concerns Wide vs Narrow and Delay time. Let’s choose a wide angle sensor with a delay time of 5 minuets.

    Figure 28 – Now we have to determine if we want are fixtures to turn-off at the last occupancy event trigger or if we want them to dim to a pre-determined output, i.e. 40%. I’m going to select ‘Full Off”.

    Figure 29 – The last two parameters to consider are grouping and photocontrol override. In this example I’m selecting no grouping which means each fixture is going to act independently. Also, we are going to enable photocontrols where applicable, i.e. daylight harvesting zones. Photocontrols are usually programmed to override control strategies so it’s important that you have them applied only where necessary.

    So, to recap, our lighting systems controls will consist of a lighting schedule, fixtures with embedded sensors (occupancy and photocontrol), fixture based control instead of grouping, and our fixtures will turn ‘Full Off’ rather than dimming to a percentage.

    There are defined communication channels of various detail to interact with LLLCs but occupants have to be informed that these methods exist, but also, become familiar with these new channels of communication. Selecting LLLC products and settings for maximum energy savings has the potential to adversely impact occupant satisfaction. While we at the IDL actively promote EEMs we have a motto, energy efficiency but not at the cost of occupant comfort.

    For example, a previous project that I worked on for LLLCs an occupant commented on the grouping and last occupancy parameters in a negative manner describing it as creepy or scary. The occupant was referring to the LLLCs being grouped individually as well as when last occupancy was detected they would go to full off. So when this occupant would leave their studio space to go home, visit another floor, or use the restroom they were uncomfortable moving through the building with all or majority of the lights being off. Of course, the lights would turn on when the occupancy sensor was triggered but the emotional perception of the space was changed by using these parameters. The building, before being retrofitted, did have lighting control strategies but were zoned (area based). Therefore, the occupant was ‘used to’ having all the lights turn on when entering a space rather than individual fixtures turning on per occupancy demand. This has been a recurring issue with almost all LLLCs projects I have worked on or consulted for, occupants unfamiliar with the new lighting technology resent or reject LLLCs. However, in my opinion I think this has to do with the lack of established methods of communication between building occupants and LLLC technology.

    The charts below, figures 29 – 32, demonstrate the impact parameters settings can have on energy use when compared to the base case energy use of no controls. Figure 29 shows the difference between choosing a wide angle versus a narrow angle for your occupancy. We can infer that a narrow field of view results in fewer occupancy action triggers, but also, that we are reducing false occupancy action events from occupants walking by. In addition, we don’t have to be concerned about lights not turning on with a narrow field of view since all our lights have an occupancy sensor embedded within. Furthermore, it also compares the impact of the delay time parameter which is easily understood that if you leave a light on longer it will use more energy. The next chart, figure 30, illustrates the impact of combining our occupancy sensor type with grouping parameters. Groups of one in this instance refer to fixtures being ungrouped. As we increase our fixture grouping from individually, one, to groups of eight our savings from selecting a narrow field of view occupancy sensor are erased. This is because as we increase our group size we are interfering with lights that could remain off but are instead being turned on as a result of being tied to a group’s occupancy trigger event rather than an individual occupancy trigger event. The next chart, figure 31, examines the impact of dimming or full-off for lights when their last occupancy event is triggered disregarding grouping size. The main reason for demonstrating the differences for this parameter refers to my example from earlier about an occupant being uncomfortable with lights turning on by demand only. You can program LLLCs so that instead of turning off, lights can instead be dimmed. This can be a transition or compromise with building occupants who are unfamiliar with LLLCs, however, you will not be able to completely capture the full amount of potential energy savings by utilizing the dimmed parameter rather than turning the lights off. As mentioned this can be used for a transitional period where building occupancy can become accustomed to lighting via demand control strategies and may eventually embrace the new technology.

    The amount of data required for processing analysis as well as impact on energy use are why you see savings or recommended settings presented in a range rather than a fixed value and LLLCs takes this and makes it a fundamental pillar for lighting technology. Because the parameters and saving are given in ranges then the lights themselves should be able to operate within those ranges. Which means that you can reprogram your lights to fit the needs of the space as it changes over time or for temporary demands of the space. More so, this lighting technology is pushing personalization of lights where occupants can have lighting in their ‘area’ meet their needs as well as standards of comfort. Occupants can use applications on their phone or a website to adjust lighting settings in addition to conventional light switches. Furthermore, because these are ‘smart’ application interacting with the lighting system data can be harvested to improve user interface and features, such as, developing a lighting preference profile for individual occupants. Occupants would be able to ‘carry’ their lighting preference with them throughout the building and eventually wherever there is smart lighting infrastructure.

    Figure 30 - With a delay time of 20 minutes and not grouping the troffers, LLLCs with wide fields of view reduce energy use by 40% compared with the base case. On Average, the narrow field of view reduces energy use by 18% compared with the wide field of view.

    Figure 31 - A group size of one means that the LLLCs are ungrouped which is the default option. When pairing LLLCs we see an increase in energy use by 10% for wide field view and 18% for the narrow field of view when comparing against ungrouped LLLCs. When LLLCs are in groups of 8 we see energy use increases by 25% for wide field of view and by 42% for narrow field of view when compared with ungrouped LLLCs.

    Figure 32 - Dimming the lighting uses 17% more energy than if the LLLCs were turned off completely, with a 20 minute delay time.

    Figure 33 - The lower potential energy savings from using zone controls in open offices compared to other commercial spaces is likely due to higher occupancy rates of open offices. These results illustrate that using LLLCs rather than zone controls in open office can result in energy savings comparable to using zone controls in spaces with lower occupancy rates.

    LLLCs potential energy savings are based on scale and scope of parameters for a building meaning that bigger projects will more readily utilize savings from this lighting technology, but also, its flexible programming. Now that we have seen the theoretical cost and implied energy savings let’s apply this to a case study. The GEM building is a 24,000 square foot office building constructed in the 1970s. It has since been remodeled and is intended to serve as an arts hub for the Boise Bench, greater Boise, and the treasure valley.

    The GEM Building and Daylight Model of GEM Building

    GEM Building first floor with Daylight harvesting zones drawn.

    The image on the left is of the first floor gallery space and the image on the right is an open studio.

    The original fixtures (382) were T12 linear fluorescent with each fixture using 160 Watts and an efficacy of 78.75 lumens per Watts. The retrofit installed 248 LLLCs with each fixture using 35 Watts and an efficacy of 131 lumens per Watts. The reduction in the overall number of fixtures is due to the building being over-lit, but also, LEDs are more efficient than fluorescent fixtures. Furthermore, the reduction resulted in a 85.7% lower LPD of .4 Watts per square foot saving an annual operation cost of 52,260 kWh. As mentioned earlier, it becomes more difficult to implement control strategies with a small kWh pool because the amount of savings is based on reducing that new pool of kWh. Therefore, return on investment takes longer and the cost-benefit analysis starts to shift in favor of little to only required controls. One reason the client was on-board with installing LLLCs is the flexibility of the technology. The repurposed building could be repurposed again or only a portion of it, for example, remodeling a floor for a new tenant. The chart below, figure 33, shows a 21% savings opportunity applied to both fixture scenarios for the GEM building, T-12 vs LED. We can clearly see that applying control strategies for the entire building with T-12s will allow us to save more kWh hours than if we used LEDs.

    Figure 34

    Figure 35

    However, with that framing it can be considered a short term gain but long term loss since the T-12’s consume significantly more kilo-watts than the LEDs or LLLCs. Now, let’s look at how individual parameters affect the cost of our lighting installation as well as any operational cost impacts.

    Table 1

    Table 2

    Tables 1 and 2 above assume the following parameters for narrow and wide field of view, a time delay of five minutes, grouping eight fixtures, and dimming to twenty percent instead of turning off. The values in Table 1 are expressed in total cost, left, and kWh, on the right. We can correlate the change in cost with the impact the parameters have on kWh hours to see the positive and negative change in cost. Even though some of our parameter settings increase the cost of the lights we still come out with a net positive range of $1.55 to $1.86. Therefore, it’s possible to achieve energy efficiency and occupant comfort with a net positive gain. Table 2, above, breaks down each parameter cost impact and how it effects the cost and rate of return in years.

    The next two figures, 36 & 37, demonstrate the same concept but with the parameters changed to a delay time of fifteen minutes, groups of four, but keeping the dimming to twenty percent rather than full-off. These new settings for the parameters results in a net gain of $3.88 to $4.20, however, the rate of return for either scenario is relatively far off. For instance, in Table 2 the rate of return is between nine and eleven years. With this in mind I want to reevaluate the cost to include hardware and incentives so we can have a better understanding of the rate of return.

    Figure 36

    Figure 37

    As shown in figure 22 we see the average cost for LLLCs fixtures for three different tiers and we know the number of fixtures installed in the GEM building, therefore, we can estimate the cost of hardware as well as the operational cost. The GEM building has a total of 26,395 kWh using LLLCs with a total area of 19,186 square feet and today it would qualify for Idaho Power incentives L1 and L3. Under 2015 IECC there is an allowable LPD of .82 and our LPD is approximately 0.47 which is 42.68% better than code. The L1 incentive can be calculated at approximately $5,755.80 when considering $0.30 per square foot. Additionally, the L3 incentive specifically for LLLCs is calculated at $1,441.17 where the incentive is paid out at 21% of saved kWh or 21% of 26,395 kWh, the total amount of kWh from the lighting system. The GEM building’s total applicable Idaho Power incentive for only the lighting system is approximately $7,196.96. In the following figures, 38 & 39, we can compare the cost of hardware between standard LEDs and LLLCs. In figure 38 we compare the cost of installing and operating LLLCs for the minimum, average, and maximum assumed cost for that fixture type. The blue bars on the chart represent the cost of just the hardware. Then we apply the one-time savings from operational cost of upgrading to LEDs fixture type, shown in orange, and lastly we examine the reduction of cost through Idaho Power incentives shown in gray.

    Figure 38

    Figure 39

    In the next chart, figure 39, we examine the same context of the GEM building but instead we only consider if the fixture type was a standard LED. We can see that a standard LED lighting system would cost less, however, our rate of return ends after the installation as well as any non-energy benefits an LLLCs lighting system provides. This tends to be where LLLCs are removed from consideration as a viable fixture type, so why then did this building choose to use LLLCs? The main reason is due to the flexibility of LLLC technology was applicable to the long term use of the building where various tenants, studios, and gallery installations are going to have varying demands of lighting for the building’s intended use as a cultural arts hub.

    Building projects whose typology falls under low occupancy or high variable occupancy rates as well as buildings with large square footage will benefit the most from LLLCs in regards to energy efficiency and programming flexibility. Therefore, lets perform a similar analysis like we did on the GEM building but for a standard 40,000 square foot building. In figures 39 & 40 we keep the same rate of cost for LLLCs and LEDs, however, because our building has doubled in area we are going to need more fixtures. So, in this example we are assuming a 2’X4’ troffer with a uniform illumination goal of 40 foot-candles. Therefore, we will need approximately 342 fixtures at 34 Watts per fixture which results in 11,628 Watts. Using our operating hours from earlier of 2,610 hours gives us approximately 30,349 kWh. Now we need to consider the kWh for using T-12 fixtures and to do this we use the DOE average from CBECS where 7 kWh per square foot at 40,000 square feet gives us an estimation of 280,000 kWh. With this we can estimate the one-time operational savings from upgrading the fixtures in the 40,000 square foot building from compact fluorescent to LEDs. I say one-time even though this can be continuously because those savings can be considered “free money” that can be easily allocated elsewhere since it is derived from operation savings. This assumption is observed for both scenarios when considering payback or return on investment in terms of years.

    Figure 40

    Figure 41

    In table 3 we can observe the impact of doubling the size of building has on our rate of return. Looking at the minimum total cost assumptions we see that our payback is reduced by eight years and the maximum payback is reduced by four years. This is to demonstrate the importance of considering savings and incentives during the project’s design rather than after bid or during commissioning processes. More so it is easier to present the argument for choosing LLLCs over standard LEDs if your project is an ideal candidate for using this lighting technology and then identifying all the potential non-energy benefits.

    Table 3

    As energy codes and energy efficiency goals become more demanding, LLLCs create flexibility for space utilization and improved occupant comfort while providing controls for reducing energy demand for lighting as well as the potential for additional savings through integration with other building systems, such as, building automation/management systems. Furthermore, LLLCs allows building owners or facilities managers more data collection to inform programming choices, such as scheduling for occupancy. In addition, allows occupants to adjust the output of individual lights or groups of lights to a set level or space conditions, for energy savings and occupant comfort. Lastly, LLLCs are a lighting technology that is designed to adapt to changes over time where a building or spaces within a building may need to be repurposed from the original design intent providing energy efficiency benefits as well as non-energy benefits.

    In conclusion, LLLCs are an investment in the longevity of a building’s applicable use. While many features or capabilities of LLLCs may not be used upfront it is resilient in its capacity to adapt to changes over time and will facilitate change easier through both economic, energy, and non-energy or comfort applicability for building owners, facilities managers, and occupants.

    Dylan Agnes

    After earning a Bachelor of Science degree in Architecture from the University of Idaho, Moscow, Dylan studied the science and engineering of building design by completing a Master's degree in architecture. As a student he worked at the Integrated Design Lab and gained hands-on experience in the practice of Integrated Design. As an IDL Research assistant, Dylan worked with both the architectural and engineering side of integrated design, providing a broader opportunity to cross over fields of study. He started working on real world projects at the Lab in the spring of 2015 and, graduated with a Master's of Architecture in Fall of 2017 with an emphasis in urban planning and net-zero/energy efficiency building design. Shortly after graduation Dylan began working as a Research Assistant at the IDL and has since been working on a wide range of projects from Energy Modeling to Daylighting Design.


  • (ICC), International Code Council. “2015 International Energy Conservation Code (IECC): ICC Digital Codes.” 2015 INTERNATIONAL ENERGY CONSERVATION CODE (IECC) | ICC DIGITAL CODES, January 2016.
  • (ICC), International Code Council. “2018 International Energy Conservation Code (IECC): ICC Digital Codes.” 2018 INTERNATIONAL ENERGY CONSERVATION CODE (IECC) | ICC DIGITAL CODES, August 2017.
  • E3T. “Advanced Lighting Control Systems.” Advanced Lighting Control Systems. Washington State University, 2013.
  • Enscoe, Abby, and Fancis Rubinstein. “Saving Energy with Highly-Controlled Lighting in an Open-Plan Office.” Lawrence Berkley National Laboratory, 2010.
  • Kisch, Teddy, and Kate DoVale. “2020 Luminaire Level Lighting Controls Incremental Cost Study.” NEEA, January 7, 2021.
  • Mahic, Alan, Jeff Kline, Dale Northcutt, and Kevin Van Den Wymelenberg. “Luminaire Level Lighting Controls Replacement vs Redesign Comparison Study.” NEEA, September 3, 2020.
  • Parrett, David. “Fixture vs. Area-Based Control: the Pros and Cons.” Cortet by CEL, July 9, 2018.
  • Rubinstein, Abby, and Francis Rubinstein. “Saving Energy with Highly-Controlled Lighting in an Open-Plan Office.” Leukos, July 2010.
  •, Energy Solutions. “Luminaire Level Lighting Controls FAQ.” Seattle: Seattle City Light, n.d.
  • Snyder, Jeremy. “Energy-Saving Strategies for Luminaire-Level Lighting Controls.” Troy: Lighting Research Center, January 10, 2018.
  • Technology Application Guide, Zero Net Energy. “Luminaire Level Lighting Control.” new building institute , November 4, 2014.
  • Warmerdam, Kevin, Ashish Pandharipande, and David Caicedo. “Connectivity in IOT Indoor Lighting Systems with Visible Light Communications.” IEEE Xplore. Philips Research, 2015.